Malaria remains a significant global health concern, particularly in tropical and subtropical countries. Indonesia, one of nine malaria-endemic countries in tropical Southeast Asia, has set a goal to eliminate malaria by 2030 (WHO, 2015). The country has made remarkable progress in malaria control, reducing the incidence from 1.8 million cases in 2011 to 811,636 cases in 2021 (WHO, 2023). In 2022, it was estimated that only 6.4% of Indonesia’s population was at high risk of malaria with 8.2% still living in active foci areas (WHO, 2023). As malaria-endemic areas in Indonesia shrink and become more localized, ongoing efforts to control and monitor the disease are crucial to containing its transmission.

Measuring malaria transmission patterns is important for effectively targeting control strategies and evaluating their impact after implementation. One method to estimate transmission involves assessing the malaria-specific immune responses in local populations, which serve as indicators of exposure to infection (van den Hoogen and Drakeley, 2017). Serological markers offer an alternative to traditional surveillance methods, allowing for the effective measurement of malaria transmission through the detection of antimalarial antibodies developed in response to the parasite antigens (Zakeri et al., 2016; Idris et al., 2017a). In low malaria transmission settings, this validated approach not only serves as a proxy to conventional methods but also provides greater sensitivity and reliability (van den Hoogen and Drakeley, 2017). As such, it is a valuable tool for guiding tailored malaria control programs and monitoring changes in transmission following intervention (Cook et al., 2010; Dewasurendra et al., 2017; Idris et al., 2017b; Macalinao et al., 2023). Numerous studies conducted in low-endemicity regions have demonstrated positive outcomes in using seroepidemiological analysis in determining the malaria burden, leading to improved public health policy planning (Rosas-Aguirre et al., 2013; Zakeri et al., 2016; Idris et al., 2017a; Keffale et al., 2019; Rahim et al., 2022; Rahim et al., 2023; Pinedo-Cancino et al., 2024).

Investigating the application of serological metrics in order to understand the historical patterns of malaria transmission in a population is essential in Indonesia. Whilst several seroepidemiological studies have been conducted in provinces in Indonesia with a very low prevalence of Plasmodium falciparum and Plasmodium vivax namely Central Java (Bretscher et al., 2013), Lampung (Supargiyono et al., 2013), Aceh (Surendra et al., 2019) and Yogyakarta (Surendra et al., 2020), no such study has been carried out in North Sumatera Province where the prevalence of Plasmodium knowlesi and multispecies of human malaria have been reported (Lubis et al., 2017; Nainggolan et al., 2022). In the present study, antibody responses to P. falciparum and P. vivax blood-stage antigens apical membrane antigen 1 (AMA-1), and merozoite surface antigen-1₁₉ (MSP-1₁₉) were measured to assess malaria exposure and transmission in North Sumatera Province.

This cross-sectional study, nested within a prior epidemiological survey on submicroscopic malaria (Nainggolan et al., 2022), analyzed antibody responses to AMA-1 and MSP-1₁₉ of P. falciparum and P. vivax in samples collected from the population in Langkat district, North Sumatera Province, Indonesia. The serological outcomes revealed similarities in the seroprevalence of both species and a relationship between age groups and seroprevalence rates. The study also identified higher transmission levels of P. falciparum compared to P. vivax based on SCR and found that spending nights in the forest was a risk factor associated with P. vivax exposure. These findings could help inform malaria elimination efforts in Indonesia and support the potential integration of seroepidemiological methods into routine elimination programs.

The antigens AMA-1 and MSP-1₁₉ were selected because they are present in both species, represent the erythrocytic stage of the parasites, and have been widely used as markers of exposure in the past. The analysis of antibody responses among the study participants revealed a similar seroprevalence rate of 20.7%, with individuals responding to at least one antigen from either P. falciparum or P. vivax. Although no other serological surveys had been conducted previously in Langkat district, these rates were relatively higher compared to those found in similar epidemiological settings in Sumatera, such as a population survey in Aceh Province, Indonesia, in 2013 (6.9% for P. falciparum and 2%, for P. vivax) (Surendra et al., 2019). Nevertheless, these findings are inconsistent with reports from other countries with P. falciparum–P. vivax co-endemicity in the Southeast Asian region where antibodies against P. falciparum antigens were found to be higher than P. vivax antigens such as Malaysia (58% and 10%) (Rahim et al., 2023), Thailand (79% and 40%) (Rahim et al., 2022), Vietnam (38% and 31%) (San et al., 2022), and Myanmar (30% and 14%) (Edwards et al., 2021). The observed difference in seroprevalence rates could be attributed to the fact that antibody responses reflect cumulative exposure events, including past asymptomatic submicroscopic infections (Idris et al., 2017a; Macalinao et al., 2023), and the relatively long half-life of responses against Pf/PvAMA-1 and Pf/PvMSP-1₁₉ antigens generated after exposure (San et al., 2022). Additionally, the historical co-dominance of P. falciparum and P. vivax in the study area with similar prevalence of 13.8% and 13.6%, respectively (Lubis et al., 2017), may have contributed to the similar seroprevalence rates observed for both species in the population, highlighting the need for targeted malaria elimination strategies in Indonesia.

Seroprevalence reflects cumulative malaria exposure and modelling changes between seroprevalence and age (i.e. SCR), can be used to estimate transmission intensity in a population (Idris et al., 2017a; van den Hoogen and Drakeley, 2017). In this study, SCRs estimated from the age-adjusted seroprevalence curves for P. falciparum antigens were higher than the ones for P. vivax antigens, reflecting the more intense transmission of the former species in Langkat district. These differences in SCR estimates are likely to reflect the different ecological factors that affect malaria exposure and the acquisition of immunity to malaria in the area (Cunha et al., 2014). Unfortunately, data on the ecology of malaria parasitism in this survey were not recorded to enable the testing of these hypotheses. Additionally, different transmission patterns for falciparum and vivax as evidenced by data are likely to reflect the actual difference in transmission of the two species observed over the years. Furthermore, the higher seroconversion rate of P. falciparum compared to P. vivax may be due to its more frequent symptomatic infections, higher transmission intensity, and possibly longer persistence of antibodies post-infection (Kattenberg et al., 2018; Herman et al., 2023). P. falciparum also typically causes more severe disease, prompting a stronger and more detectable immune response over time (van den Hoogen et al., 2020).

The estimated SCR for the AMA-1 was higher than that of MSP-1₁₉ for both P. falciparum and P. vivax in Langkat District, aligning with results from other seroepidemiological studies (Cook et al., 2010; Tusting et al., 2014; Wong et al., 2014; Idris et al., 2017b; Kwenti et al., 2017; Surendra et al., 2019; van den Hoogen et al., 2020). The differences in transmission estimates between AMA-1 and MSP-1₁₉ may be attributed to variations in seroconversion and reversion rates, which are potentially influenced by differences in immunogenicity, subclass-dependent half-life, and antigen polymorphism (Badu et al., 2012). AMA-1, being more immunogenic and associated with higher antibody titers than MSP-1₁₉, likely exhibits faster seroconversion and seroreversion rates (Drakeley et al., 2005), which might also explain the observations. Furthermore, the absence of an age-related trend in seroprevalence for the AMA-1 and MSP-1₁₉ antigens of both P. falciparum and P. vivax in the present study limits their value when analyzed using the current modelling approach. Alternative serological estimates of malaria transmission intensity, such as the antibody acquisition model (Yman et al., 2016) and the unified mechanistic model (Kyomuhangi and Giorgi, 2021) can enhance the precision of transmission estimates.

Multivariate regression analyses identified that for P. falciparum, no variables were significantly associated with seropositivity in the adjusted model. However, spending nights in the forest was associated with P. vivax seropositivity. This observed association could be due to the distinct ecological and behavioral patterns of the vectors that transmit P. vivax, which are often more exophagic (outdoor-biting) and exophilic (outdoor-resting) compared to those transmitting P. falciparum (Baird, 2013). A recent study in Sumatera showed that predominant P. vivax vectors, such as Anopheles dirus (Aceh Province) and Anopheles kochi (North Sumatera Province), thrive in forested environments where they have greater access to humans sleeping outdoors or in forested areas (Permana et al., 2023), which could lead to higher exposure risk. This is consistent with the reported risk factors from a previous study, where 59.4% of individuals had forest-associated occupation, and 75.7% had a history of forest visits (Nainggolan et al., 2022). In contrast, P. falciparum vectors are generally more endophagic (indoor-biting), reducing the likelihood of transmission in forest settings (Cohen et al., 2012). The resilience of P. vivax to low-density environments and its ability to persist in temperate climates further increases its transmission potential in forested regions (Battle et al., 2012; Yman et al., 2016). Additionally, P. vivax hypnozoites can cause relapses, maintaining seropositivity even with intermittent exposure (Flannery et al., 2022).

This study had several limitations. The most significant was the insufficient sample size, which inevitably reduced the precision of the current SCR estimates (Sepúlveda et al., 2015). Additionally, the disproportionate sampling and overrepresentation of adult participants may have led to results that are not fully representative of the study population. This oversampling likely occurred due to the timing of surveys, as most were conducted on weekdays when children were in school. Furthermore, the study was restricted to a small geographical area in Langkat district, a region with very low malaria endemicity (Lubis et al., 2017; Nainggolan et al., 2022). This limitation raises concerns about the generalizability of the findings to other areas of Indonesia or beyond. Lastly, the study utilized only two recombinant antigens for each Plasmodium species. While these antigens are considered long-term markers of transmission (Drakeley et al., 2005), individual variations in immune responses to different parasite antigens suggest that using a broader range of antigens, including short-term markers, might improve the identification of seropositive individuals (Kerkhof et al., 2016). Despite these limitations, the study contributes to the collective body of knowledge on malaria antibodies and lays the groundwork for the potential future use of this tool in Indonesia, which is especially relevant for other countries also aiming for malaria elimination.

This study assessed antibody responses to AMA-1 and MSP-1₁₉ antigens of P. falciparum and P. vivax in Langkat District, Indonesia, revealing similarities in seroprevalence and significant age-related variations in seropositivity rates. The findings indicate higher transmission levels of P. falciparum compared to P. vivax, aligning with global patterns but diverging from observations in co-endemic settings in Southeast Asia. The association between spending nights in the forest and increased P. vivax seropositivity highlights the role of vector behavior and ecological factors in transmission dynamics. These insights emphasize the need for targeted interventions tailored to specific populations with varying demographics and risk factors, as well as the integration of seroepidemiological tools into malaria elimination strategies in Indonesia. By understanding the distinct transmission patterns and associated risk factors, tailored approaches can be developed to address the specific challenges of P. falciparum and P. vivax elimination.